We use $ref in the schema to allow us to specify multiple type, eg
command, it can be a string or a list of strings.
It required some extra parsing to retrieve a helpful type to display
in our error message rather than 'string or string'. Which while
correct, is not helpful. We value helpful.
Signed-off-by: Mazz Mosley <mazz@houseofmnowster.com>
When a schema type is set as unique, we should display the validation
error to indicate that non-unique values have been provided for a key.
Signed-off-by: Mazz Mosley <mazz@houseofmnowster.com>
The validation message was confusing by displaying only 1 level of
property of the service, even if the error was another level down.
Eg. if the 'files' property of 'extends' was the incorrect format,
it was displaying 'an invalid value for 'extends'', rather than
correctly retrieving 'files'.
Signed-off-by: Mazz Mosley <mazz@houseofmnowster.com>
Unfortunately the way that jsonschema is calling %r on its property
and then encoding the complete message means I've had to do this
manual way of removing the literal string prefix, u'.
eg:
key = 'extends'
message = "Invalid value for %r" % key
error.message = message.encode("utf-8")"
results in:
"Invalid value for u'extends'"
Performing a replace to strip out the extra "u'", does not change the
encoding of the string, it is at this point the character u followed
by a '.
Signed-off-by: Mazz Mosley <mazz@houseofmnowster.com>
While it was intended as a positive to be stricter in validation
it would in fact break backwards compatibility, which we do not
want to be doing.
Consider re-visiting this later and include a deprecation warning if
we want to be stricter.
Signed-off-by: Mazz Mosley <mazz@houseofmnowster.com>
If a container is in the process of being removed, or removal has
failed, it can sometimes appear in the output of GET /containers/json
but not have a 'Name' key. In that case, rather than crashing, we can
ignore it.
Signed-off-by: Aanand Prasad <aanand.prasad@gmail.com>
Rather than implement the logic a second time, use docker-py
split_port function to test if the ports is valid.
Signed-off-by: Mazz Mosley <mazz@houseofmnowster.com>
Move validation out into its own file without causing circular
import errors.
Fix some of the tests to import from the right place.
Also fix tests that were not using valid test data, as the validation
schema is now firing telling you that you couldn't "just" have this
dict without a build/image config key.
Signed-off-by: Mazz Mosley <mazz@houseofmnowster.com>
jsonschema provides a rich error tree of info, by parsing each error
we can pull out relevant info and re-write the error messages.
This covers current error handling behaviour.
This includes new error handling behaviour for types and formatting of
the ports field.
Signed-off-by: Mazz Mosley <mazz@houseofmnowster.com>
We validate the config against our schema before a service is created
so checking whether a service name is valid at time of instantiation of
the Service class is not needed.
Signed-off-by: Mazz Mosley <mazz@houseofmnowster.com>
Define a schema that we can pass to jsonschema to validate against the
config a user has supplied. This will help catch a wide variety of common
errors that occur.
If the config does not pass schema validation then it raises an exception
and prints out human readable reasons.
Signed-off-by: Mazz Mosley <mazz@houseofmnowster.com>
- Better method names.
- Environment variable syntax in volume paths, even when a driver is
specified, now *will* be processed (the test wasn't testing it
properly). However, `~` will still *not* expand to the user's home
directory.
Signed-off-by: Aanand Prasad <aanand.prasad@gmail.com>
* Add support for volume_driver parameter in compose yml
* Don't expand volume host paths if a volume_driver is specified
(i.e., disable compose feature "relative to absolute path
transformation" when volume drivers are in use, since volume drivers
can use name where host path is normally specified; this is a
heuristic)
Signed-off-by: Luke Marsden <luke@clusterhq.com>
When an image declares a volume such as `/var/lib/mysql`, and a Compose
file has a line like `./data:/var/lib/mysql/` (note the trailing slash),
Compose creates duplicate volume binds when *recreating* the container.
(The first container is created without a hitch, but contains multiple
entries in its "Volumes" config.)
Fixed by normalizing all paths in volumes config.
Signed-off-by: Aanand Prasad <aanand.prasad@gmail.com>
It doesn't do much other than cause the remainder of the test suite to
generate lots of junk output.
Signed-off-by: Aanand Prasad <aanand.prasad@gmail.com>
Rather than creating a docker client within each test, create one
at setup and make it accessible to the whole class.
Signed-off-by: Mazz Mosley <mazz@houseofmnowster.com>
When specifying a log_driver you want to specify some options for
the logger as per the docker run --log-opt option. The logger
options are key value pairs.
Signed-off-by: Mazz Mosley <mazz@houseofmnowster.com>
By allowing the memswap_limit option to be defined we also need to
check that mem_limit is set, you can't have swap without a limit.
Signed-off-by: Mazz Mosley <mazz@houseofmnowster.com>
Also warn the user about the one-off containers in the standard error
message about legacy containers.
Signed-off-by: Aanand Prasad <aanand.prasad@gmail.com>
This top level function is a test helper, so I've moved it into the
config_test file and updated accordingly.
Signed-off-by: Mazz Mosley <mazz@houseofmnowster.com>
When building test data using make_service_dict, we need to include
working_dir as it is core to some of the functionality of
ServiceLoader.
Signed-off-by: Mazz Mosley <mazz@houseofmnowster.com>
If you have an alternate YAML file with different services defined,
containers for those services will be shown in `docker-compose ps` even
if you don't pass that file in.
Furthermore, `docker-compose rm` will claim that it's going to remove
them, but actually won't.
Signed-off-by: Aanand Prasad <aanand.prasad@gmail.com>
When specifying the 'file' key to a value of it's own name, test
that this works and does not cause a false positive circular reference
error.
Signed-off-by: Mazz Mosley <mazz@houseofmnowster.com>
This refactoring allows us to raise an error when there is no
'file' key specified in the .yml and no self.filename set. This
error was specific to the tests, as the tests are
the only place that constructs service dicts without sometimes
setting a filename.
Moving the function within the class as well as it is code that
is exclusively for the use of validating properties for the
ServiceLoader class.
Signed-off-by: Mazz Mosley <mazz@houseofmnowster.com>
Split them out into individual validation tests so it is clearer
to see what is going on and to enable adding further validation
tests.
Signed-off-by: Mazz Mosley <mazz@houseofmnowster.com>
If the 'file' key is not set in the extends_options dict then we
look for the 'service' from within the same file.
Fixes this issue: https://github.com/docker/compose/issues/1237
Signed-off-by: Mazz Mosley <mazz@houseofmnowster.com>
This commit adds environment variable parsing to the container side
of the volume mapping in configs. The common use case for this is
mounting SSH agent sockets in a container, using code like:
volumes:
- $SSH_AUTH_SOCK:$SSH_AUTH_SOCK
environment:
- SSH_AUTH_SOCK
Signed-off-by: Jeff Kramer <jeff.kramer@voxmedia.com>
As VALID_CHARS is shared with project names, these chars are also
now allowed within project names.
Signed-off-by: Mazz Mosley <mazz@houseofmnowster.com>
These tests were indeed raising a config error, but not for the reason
intended/tested for. I've added in the image name so the config error
raise is correctly testing the Service name.
Signed-off-by: Mazz Mosley <mazz@houseofmnowster.com>
Add a test to make sure NoSuchService is raised if
a bogus service name is given to 'docker-compose logs'.
Signed-off-by: Dan O'Reilly <oreilldf@gmail.com>
I think the original intention of the original test was the check that
different signals work, I think. This does this -- it sends a signal
that doesn't cause the container to stop.
Closes#759. Replaces #1467.
Signed-off-by: Ben Firshman <ben@firshman.co.uk>
This fixes a bug where migration would fail with an error if a
downstream container was migrated before its upstream dependencies, due
to `check_for_legacy_containers()` being implicitly called when we fetch
`links`, `volumes_from` or `net` dependencies.
Signed-off-by: Aanand Prasad <aanand.prasad@gmail.com>
When an upstream dependency (e.g. a db) has a container but a downstream
service (e.g. a web app) doesn't, a web container is not created on
`docker-compose up`.
Signed-off-by: Aanand Prasad <aanand.prasad@gmail.com>
- One-off containers were included in the warning log messages, which can
make for unreadable output when there are lots (as there often are).
- Compose was attempting to recreate one-off containers as normal
containers when migrating.
Fixed by implementing the exact naming logic from before we used labels.
Signed-off-by: Aanand Prasad <aanand.prasad@gmail.com>
- Rename `migration` module to `legacy` to make its legacy-ness explicit
- Move `check_for_legacy_containers` into `legacy` module
- Fix migration test so it can be run in isolation
Signed-off-by: Aanand Prasad <aanand.prasad@gmail.com>
linting...
six.string_types
list-of-strings in examples
disallow extra_hosts support for list-of-dicts
A more thorough sets of tests for extra_hosts
Provide better examples
As per @aanand's [comment](https://github.com/docker/compose/pull/1158/files#r28326312)
I think it'd be better to check `if not isinstance(extra_hosts_line,
six.string_types)` and raise an error saying `extra_hosts_config must be
either a list of strings or a string->string mapping`. We shouldn't need
to do anything special with the list-of-dicts case.
order result to work with assert
use set() instead of sort()
Signed-off-by: CJ <lim@chernjie.com>
Does not change directory to the parent with the compose-file found.
Works like passing '--file' or setting 'COMPOSE_FILE' with absolute path.
Resolves issue #946.
Signed-off-by: Aleksandr Vinokurov <aleksandr.vin@gmail.com>
* This fix introduces one side-effect: the build parameter is now
validated early, when the service dicionary is first constructed.
That leads to less scary stack traces when the path is not valid.
* The tests for the changes introduced here alter the fixtures
of those (otherwise unrelated) tests that make use of the 'build:'
parameter)
Signed-off-by: Moysés Borges Furtado <moyses.furtado@wplex.com.br>
Karol Duleba
Mazz Mosley
mnowster
Aanand Prasad
Ben Firshman
Veres Lajos
Yuval Kohavi
Luke Marsden
Daniel Nephin
Jeff Kramer
Sébastien Gruchet
Travis Thieman
Vincent Demeester
Dan O'Reilly
André Martins
Todd Whiteman
lsowen
delbert@umn.edu
CJ
xuxinkun
Timothy Van Heest
Thomas Desvenain
Sam Wing
Michael Chase-Salerno
Aleksandr Vinokurov
Steven Dake
Joseph Page
Moysés Borges
akoskaaa